Visual Summarization of Temporal Event Sequences Using MDL Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The display of large sets of event sequences in modern computing systems often results in visual clutter, overwhelming users and reducing the usefulness of the data for analysis, as existing methods fail to effectively balance reducing clutter with increasing information content.
Innovation Solution
The implementation of a minimum description length (MDL) optimization process for generating graphical depictions of event sequences, which includes simultaneous sequence clustering and pattern extraction, tolerant to noise, and provides a visual analytics framework with multiple levels-of-detail for interactive data exploration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If large sets of event sequences are displayed in graphical form, then the information content is increased, but visual clutter increases and overwhelms users
Solution Approach 1:
The patent segments event sequences into clusters based on similarity, where each cluster represents a group of sequences with common patterns. This segmentation reduces visual clutter by grouping related events together while preserving the underlying information structure, allowing users to analyze patterns without being overwhelmed by individual sequence details.
Solution Approach 2:
The patent merges similar event sequences into clustered representations, combining multiple sequences that share common patterns into a single visual group. This merging reduces the overall visual complexity while maintaining the essential information content, as the clustered display preserves pattern recognition capabilities without displaying every individual sequence.
2Object-affected harmful factors
If visual summarization is applied to reduce clutter, then visual clarity is improved, but information content may be lost
Solution Approach 1:
The patent implements a dynamic visual summarization system that adapts the level of detail based on user interaction and zoom level. At higher levels of detail, individual sequences are visible; at lower levels, clustered patterns are emphasized. This dynamic approach ensures that information is preserved when needed while reducing clutter when viewing the overall structure.
Solution Approach 2:
The patent introduces clustered pattern representations as intermediary visual elements between individual event sequences and overall summaries. These intermediaries preserve essential pattern information while reducing visual complexity, acting as a bridge that maintains information integrity without overwhelming the user with raw sequence data.
3Loss of information
If detailed event sequences are displayed, then information content is preserved, but understanding and analysis become difficult
Solution Approach 1:
The patent applies local quality by allowing users to drill down from clustered summaries to individual sequences only in regions of interest. The visual representation provides different levels of detail in different areas - clustered patterns at the overview level and detailed sequences when users focus on specific clusters, making the system adaptable to local analysis needs.
Solution Approach 2:
The patent adds a hierarchical dimension to the visual representation, organizing event sequences into multiple levels of abstraction. Users can navigate from high-level clustered patterns to detailed individual sequences by moving through different hierarchical levels, transforming a two-dimensional cluttered display into a multi-dimensional navigable structure that improves understandability.
Data Source
AI summary
A method for generating a graphical depiction of summarized event sequences includes receiving a plurality of event sequences, each event sequence in the plurality of event sequences including a plurality of events, and generating a plurality of clusters using a minimum description length (MDL) optimization process. Each cluster in the plurality of clusters including a set of at least two event sequences in the plurality of event sequences that maps to a pattern in each cluster. The pattern in each cluster further includes a plurality of events included in at least one event sequence in the set of at least two event sequences in the cluster. The method includes generating a graphical depiction of a first cluster in the plurality of clusters, the graphical depiction including a graphical depiction of a first plurality of events in the pattern of the first cluster.